Wearable ECG for Real Time Complex P-QRS-T Detection and Classification of Various Arrhythmias
| Autor: | Neha Arora, Biswajit Mishra, Yash Vora |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
medicine.diagnostic_test
Computer science business.industry 0206 medical engineering Feature extraction Approximation algorithm Pattern recognition 02 engineering and technology 020601 biomedical engineering Signal 030218 nuclear medicine & medical imaging 03 medical and health sciences QRS complex 0302 clinical medicine T wave Metric (mathematics) medicine Artificial intelligence Sensitivity (control systems) business Electrocardiography |
| Zdroj: | COMSNETS |
| DOI: | 10.1109/comsnets.2019.8711218 |
| Popis: | An ECG signal carries vital information that can be used for detecting various arrhythmia conditions. In this work, we have developed an algorithm to detect the R peaks of the ECG signal based on the Pan Tompkins Algorithm. Further, the work has been extended to a first level approximation to detect various arrhythmia conditions. Further to compute the QRS complex, the Q and S points based on the R-peaks are detected. To validate the effectiveness of the proposed algorithm, the MIT/BIH arrhythmia database is used as a source for the ECG signals and the reference for R peak annotations. The algorithm provides the metric of False Detection Rate (FDR) to be 1.289%, Sensitivity to be 99.492% and Positive Predictivity to be 99.293%. After detecting the complete QRS complex, the entire QRS complex is set to zero in order to detect the P and T waves from the signal. Hence, this work provides a specific approach of P, Q, R, S and T wave detection of an ECG signal in a real-time environment. The algorithm is further ported to a low-cost ECG monitoring wearable patch to show the effectiveness of the approach. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|